Businesses and their data analysts face the challenge of making sense of and finding patterns in the increasingly large amounts of data in the many types and formats that such businesses generate and collect. For example, accessing computer networks and transmitting electronic communications across the networks generates massive amounts of data, including such types of data as machine data and Web logs. Identifying patterns in this data, once thought relatively useless, has proven to be of great value to the businesses. In some instances, pattern analysis can indicate which patterns are normal and which ones are unusual. For example, detecting unusual patterns can allow a computer system manager to investigate the circumstances and determine whether a computer system security threat exists.
Additionally, analysis of the data allows businesses to understand how their employees, potential consumers, and/or Web visitors use the company's online resources. Such analysis can provide businesses with operational intelligence, business intelligence, and an ability to better manage their IT resources. For instance, such analysis may enable a business to better retain customers, meet customer needs, or improve the efficiency of the company's IT resources. Despite the value that one can derive from the underlying data described, making sense of this data to realize that value takes effort. In particular, patterns in underlying data may be difficult to identify or understand when analyzing specific behaviors in isolation, often resulting in the failure of a data analyst to notice valuable correlations in the data from which a business can draw strategic insight.